Dictation systems that capture speech input from a user for subsequent transcription are widely used to facilitate data entry in a variety of industries, including the health care, legal and media industries. For example, in the field of health care, doctors and other medical professionals often dictate notes regarding patients, such as in response to a patient encounter such as a patient visit, a medical procedure performed on the patient, a diagnostic analysis of patient data, etc. As another example, an attorney may dictate a memorandum or a brief (or a portion thereof) to be added to the appropriate electronic record for the associated legal matter. Independent of the field of endeavor, the dictation, once captured, is then typically transcribed (e.g., by a human transcriptionist, using automatic speech recognition, or a combination of both) and entered into or otherwise associated with the corresponding electronic record.
To satisfactorily accomplish this workflow, the correct electronic record must be identified so that corresponding transcriptions are placed in the proper location within the electronic record system. One method by which this may be achieved is by requiring the user (e.g., the dictator) to manually enter identifying information to make the necessary association between a dictation and a corresponding electronic record. For example, in the health care context, the doctor may be required to manually enter patient information, such as name, date of birth, etc. when recording the dictation by typing the information into the correct system, by dictating enough identifying information into the dictation itself, scanning a barcode on a physical file, etc.
The provided information is then extracted downstream (e.g., by a transcriptionist and/or by an ASR system) and used to associate the transcription with the electronic health record. However, this solution has the drawback of placing the burden on the physician, both in requiring the doctor to spend time providing the information and by requiring that the doctor have the appropriate information on hand. Additionally, relatively significant work may be required downstream to associate the transcription with the corresponding electronic record based on the provided information and this process is susceptible to error.
An alternative solution is to integrate the dictation system with the electronic records system so that the integrated system can automatically make the necessary associations. However, the variety of new and legacy dictation systems and electronic records systems makes doing so a difficult and potentially cost-prohibitive solution. For example, a health care facility may not want to invest the time and money in the development effort, training and/or maintenance involved in procuring and implementing an integrated solution to the above described workflow. In some circumstances, it may not even be possible to integrate a given dictation system with a given electronic records systems.